CT Denoising

CT denoising aims to improve the quality of low-dose computed tomography (CT) scans by reducing noise while preserving crucial diagnostic information. Current research heavily utilizes deep learning, particularly convolutional neural networks (CNNs) like U-Net and more recently, transformer-based architectures such as RWKV, to achieve this, exploring both supervised and unsupervised learning approaches, including self-supervised and even privacy-preserving methods. These advancements are crucial for minimizing patient radiation exposure while maintaining the diagnostic accuracy of CT images, impacting both clinical practice and medical research. A key challenge remains bridging the gap between simulated and real-world data for training robust and generalizable models.